Комментарии:
Thank you!
Ответитьlooks like it doing anything you ask as much as it can from its training data but cant do much detailed graphic description from bad things, because probably they are not in llama's the training data.
maybe i will try download other models
Is that uncensored? And how much minimum spec?
ОтветитьI love your videos. I watched some of your Stable Diffusion videos and it's great you're doing chat too. I find it hard to find a current model on Hugging face so I'll give it another try.
Ответитьram 17gb usage T_T
thanks for that thought
Sadly dont work for me. I as what is a cat, and app doesnt respond
Ответитьthank you! how do you increase prompt size? I find myself at a spot were my prompts are too long for it.
ОтветитьSystem requirements? Ghz(how many CPU cores) and RAM thanks!
ОтветитьI tried to install it and only got an error saying it couldn't open the LLM. This error occurs no matter which one I download...
ОтветитьStart chatting but nothing happens ,The UI loads fine on Windows. But when I start chatting, nothing happens.
ОтветитьI have 4GB RAM and Intel R Celeron CPU N2840.
can this work for my pc
not working, infinite loading when answering a question
ОтветитьCan we fine tune it ? With local files for instance ?
ОтветитьI'm trying to run it with vicuna and no luck! The gear is turning and Alpaca doesn't answer.
ОтветитьThank you Olivio. Your video's are amazing and VERY helpful.
ОтветитьMine dosent respond its still loading
ОтветитьSo basically the program takes only one of all those models out there? :/
Ответитьis this thing really safe? Microsoft defender is giving me a weird error not to run it. Bit defender seems to think its OK. And it probably is, but it won't let me run it unless I force it to run.
Ответитьwhen i load my model it just says "chat.exe is not responding", the app dosent crash however but it never responds to anything i say
Ответитьare there different models for different action, example a model to make programs in python, a model for example request a prompt for stable distribution
ОтветитьIs there infinite messages you can send
ОтветитьIf your grannie can't do it, it's too complicated.
ОтветитьHi!
There are a lot more files in HF repo now for model alpaca 7B... Which file are what ? Ans which ones are to use in Electron app?
There are several .bin with _1 _2 _3 are all the numbered files needed?
There are q4 q5 q8 what does it means?
What is a pickle file (.pth) ?
Thx for advices!
Bruh why is it so slow for me.
ОтветитьMy laptop is an intel i7 and have 7gb ram and 1tb ssdnvme, it is super super super super slow, for real.
ОтветитьI cannot get the BIN to open. Several programs report that the file won't extract.
ОтветитьThanks! I've been looking for a solution for running this on my Ryzen 5 2400G and... it runs!
ОтветитьBuenas lo descargue, abre la interfaz le cargo un model.y no me ande queda pensando apenas ingres la pregunta. sea cual sea, que me puede ser, tengo 32 gb ram. y 16 nucleos (mi pc). por favor alguno que me oriente gracias.
Ответить@Olivio thanks for this. Since I have an RTX2080 card with 8gb of VRAM what other program can you recommend to use with AI text models? Thanks again
ОтветитьDoes this run as GPT3.5 or GPT4? This version is not working ggml-model-q8_0.bin. I get this error: Please make sure it is a new version of the GGML model. The old one no longer works.
ОтветитьLOL
Ответитьcan u connect with different models?
like not related to alpaca?
wondering about the fine tuning process
Ответитьit's stuck loading
Ответитьpor fin se hizo mi sueño realidad tener mi propio chatgpt en mi pc, sin depender de internet, gracias maestro olivia sarikas.
ОтветитьThank you! My first thougth when ChatGPT became available was how I'd love to have it on my own laptop. You've just realized my dreams! Does this model get modified in any way when I use it locally, like supposedly ChatGPT accessed over the internet does?
Ответитьcan you hook it up through an API to another app?
ОтветитьQuick question for you. Is there a way to point Alpaca to a specific database to get it's information from? I was thinking of setting up a how to database for my company so that clients can ask how to questions on how to run our software.
ОтветитьYes! I hate it all! just give me what I want!!!!
ОтветитьI think its kinda garbage haha. most of the time it just says :"Hello! How can i help you?" and always after a german response there are hashtags. maybe i have to try a different modek
ОтветитьHoly smokes, this is awesome, it is great for role play. You can make a West World if you want and have the characters respond in different ways.
Control their feelings and emotions, they even feel guilty and lie to hide things from certain characters... lol This is amazing.
Can Alpaca Electron write computer code? I’m using the plus version of ChatGPT with GPT4 and it has really accelerated that development work. Being able to do this locally and free of charge sounds too easy to be true.
ОтветитьNot local GPT. Misleading title.
ОтветитьAs far as I know both Alpaca & Llama dataset & models don't allow commercial use & I'm not yet aware of any version that legally gets around that...
ОтветитьIs there any model available for say WordPress? or that matter HTML, CSS and JavaScrip?
Ответитьis TTS available on Alpaca?
ОтветитьThe installation is straightforward and easy, but when I run it, it is unbearable slow, seems to me it is useless. What I am doing wrong to get such bad results. Help please! Thanks!
Ответитьok be honest. How long did that list of emperors take to compute? I feel like it had to have been at least 30 minutes with it running on CPU power.
ОтветитьMy computer has basically no GPU and a decent CPU and 32GB of Ram so this is perfect
Ответить